INDEX

Explanations

technical terms related to technological processes

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 antioxid

-0.66

enced

-0.65

 pregn

-0.65

Â¢

-0.65

 Newsp

-0.64

 Webster

-0.63

berman

-0.62

 relativity

-0.62

encers

-0.61

 Flavoring

-0.61

POSITIVE LOGITS

strap

1.97

legged

1.24

stra

1.21

camp

1.17

loader

1.13

leg

0.99

lake

0.83

rom

0.82

ilage

0.81

ypes

0.80

Activations Density 0.015%

technical terms related to technological processes

No Comments

No Known Activations